Research
Evaluating Multi-Agent Coordination in Automated Code Review
An evaluation of specialized cooperating agents for reviewing pull requests, measuring how evidence gathering and role specialization affect the precision of merge recommendations.
Retrieval-Augmented Generation for Domain-Specific Question Answering
A study of retrieval strategies for grounding large language models in narrow technical domains, comparing chunking and reranking approaches for factual accuracy and citation quality.
Benchmarking FastAPI and Express for I/O-Bound APIs
A comparative benchmark of FastAPI and Express under concurrent I/O-bound workloads, examining throughput, latency tails, and memory footprint across load profiles.
PostgreSQL vs MongoDB for Hybrid Workloads
An experiment comparing relational and document databases on mixed transactional and analytics workloads, with notes on schema design trade-offs and query ergonomics.